New Nomad framework generates human mobility trajectories without target city data

Researchers have developed a novel framework called Nomad for generating human mobility trajectories without requiring data from the target city. This approach separates the learning of movement patterns from their realization on a specific city's map. Nomad utilizes a flow-matching model trained on source city data to learn transitions between Points of Interest (POIs) and then grounds these transitions onto a target city's POI map using a behavior graph and a walk mechanism. Experiments across ten cities demonstrated that Nomad outperforms existing adaptation baselines by approximately 15% in distributional fidelity and 3% in downstream utility. AI

IMPACT Enables more accurate urban planning and location-based services in data-scarce regions.

RANK_REASON Academic paper detailing a new framework for mobility generation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX
• Connected Papers
• DagsHub
• Gotit.pub
• Hugging Face
• Litmaps
• Nomad
• ScienceCast
• scite Smart Citations

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →